The purpose of this study is to test the hypothesis that autologous dendritic cells pulsed with autologous crude tumor cell lysates or melanoma peptides can be prepared and injected into melanoma patients along with IL-2 without significant toxicity. We wish to determine if treatment with melanoma tumor antigen pulsed autologous dendritic cells will increase tumor specific cytotoxic T-cell responses and induce positive skin test responses to tumor antigens. Patients eligible will be ones with metastatic disease to regional lymph nodes and can prove to be tyrosinase and/or gp100 positive. Patients are distributed into three groups. Group A will include patients without adequate tumor for preparation of tumor lysate. These patients will be tested for eligibility for the peptide pulsed autologous dendritic cell protocol, and will be included in this protocol if their tumor expresses gp100 protein and tyrosinase, and they type for HLA-A1,-A2,-A3. Group B will include patients with adequate tumor for lysate but who are not eligible for the peptide pulsed protocol because they are not HLA-A1, A2, or A3. They will be eligible for the lysate pulsed autologous dendritic cell protocol. GroupC will include patients with adequate tumor for lysate which expresses gp100 and/or tyrosinase and who type for HLA-A1, -A2, or -A3. They are eligible for randomization to receive either the peptide pulsed protocol or tumor lysate. We plan to study 100 patients for this study and at the present time there are 45 patients that have signed the consent form, have had their surgery and are approaching treatment or have already been treated. This trial is also being done at the University of Virginia with Dr. Craig Slingluff. To date they have placed 5 patients on study. We have not seen any significant adverse events. There has been skin peeling, itching and redness at time of treatment that is of short duration as well as some taste alteration. These side effects are considered to be much less than the ones that are associated with High dose Interferon and chemotherapy. There have been 5 recurrences since the protocol was started which now is over 1 year and of these 5 patients all are still alive. Our projected date for completion of this study is May of 2000. Significance: Patients with melanoma who have one lymph node involved have a 60% chance of dying secondary to melanoma and this progresses to greater that 90% if they have four or more lymph nodes involved. To date the treatments available are other phase I studies using some type of vaccine, high dose interferon, high dose IL-2, chemotherapy, surgery alone, or for local control radiation therapy. New therapies are needed for this disease.